From owner-freebsd-hackers Fri Apr 3 05:52:21 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id FAA06726 for freebsd-hackers-outgoing; Fri, 3 Apr 1998 05:52:21 -0800 (PST) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from localhost.zilker.net (jump-x2-0081.jumpnet.com [207.8.61.81]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id FAA06718 for ; Fri, 3 Apr 1998 05:52:17 -0800 (PST) (envelope-from marquard@zilker.net) Received: (from marquard@localhost) by localhost.zilker.net (8.8.8/8.8.3) id HAA27913; Fri, 3 Apr 1998 07:51:43 -0600 (CST) To: freebsd-hackers@FreeBSD.ORG Subject: Re: bcopy implementation References: <199804030436.XAA04721@dwarpal.in.oracle.com> From: Dave Marquardt Date: 03 Apr 1998 07:51:12 -0600 In-Reply-To: "Muthu"'s message of "03 Apr 98 08:21:25 +0330" Message-ID: <857m57c8tr.fsf@localhost.zilker.net> Lines: 20 X-Mailer: Quassia Gnus v0.22/XEmacs 19.16 - "Lille" Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG "Muthu" writes: > I took a look at bcopy implementation inside the kernel(FreeBSD 2.2.5). If > the number of bytes to copy is greater than 1024, it uses the floating point > unit implementation else generic bcopy. > > This kind of implementation is not used in libc. Only generic bcopy is used > in all cases. On the assumption, that the system has pentium processor, the > bcopy implementation of libc can be changed to floating point implementation > as in kernel. > > One reason of not changing implementation of bcopy in libc may be that > applications will run at different speed on different processor. On pentium it > will run faster and on 80386 it will run slow. Is there any other valid reason > of not implementing this method in libc? Well, you said it yourself. "On the assumption, that the system has pentium processor...." Well, we can't assume that, since FreeBSD runs on 386s and up. -Dave To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message